Construction of a nomogram prediction model for obstructive sleep apnea combined with hypertension via...

The purpose of this study was to explore the influencing factors of obstructive sleep apnea (OSA) combined with hypertension and construct a column chart prediction model.

A single-center retrospective case-control study was conducted to retrospectively collect data from 798 OSA patients who completed polysomnography in the Department of Otolaryngology, First Affiliated Hospital of University of Science and Technology of China, from January to December 2024. The patients were randomly divided into a training set and a validation set at a 7:3 ratio via R software correlation functions. Univariate and multivariate logistic regression was used to analyze the risk factors affecting OSA combined with hypertension, and a column chart prediction model was constructed to predict the probability of OSA combined with hypertension.

A total of 798 patients with obstructive sleep apnea syndrome were included in this study. Among them, 373 were hypertensive patients, with a hypertension incidence rate of 46.7%. Binary logistic regression analysis revealed that sex (OR = 0.467, 95% CI: 0.252-0.964), BMI (OR = 1.931, 95% CI: 1.133-3.291), and the nighttime average pulse rate (OR = 1.367, 95% CI: 1.144-1.633) were independent influencing factors of OSA with hypertension (P < 0.05). On the basis of these results, a prediction model was constructed with an area under the ROC curve of 0.667 (95% CI: 0.623-0.711). After internal validation, the area under the ROC curve of the validation group was 0.656 (95% CI: 0.588-0.724). The calibration curves of the modeling and validation groups show a good fit between the predicted results and the actual results. The DCA chart shows that the predictive model can achieve higher net returns.

This study successfully established and preliminarily validated a nomogram prediction model based on polysomnographic parameters. The model demonstrated stable discriminatory ability and consistency in both training and validation sets, which may aid in early identification of high-risk individuals with obstructive sleep apnea complicated by hypertension.